Semantic Technologies and Applications COMP5860M
John Stell
Room 9.15, School of Computing
j.g.stell@leeds.ac.uk
Lecture 7: February 2020 Ontology Development
1
Earlier
Ontology Definition(Noy & McGuiness):
An ontology defines a common vocabulary for people who need
to share information in a domain.
It includes machine-interpretable definitions of basic concepts
in the domain and relations among them. Ontologycomponents:
Classes (concepts), in hierarchy of subclasses (inheritance)
Relationships (also called properties, also called roles)
Instances (also called individuals)
Knowledge Base:
Classes+Subclasses+Relationships+Instances Example ontologies
CRM: www.cidoc-crm.org
SNOMED: https://termbrowser.nhs.uk/
Pizza: mowl-power.cs.man.ac.uk/
protegeowltutorial/resources/ProtegeOWLTutorialP4 v1 3.pdf
bioportal.bioontology.org/ontologies/ENVO
2
bioportal.bioontology.org/ontologies/ENVO
3
In this Lecture
How would you go about creating an ontology? Methodologies for ontology creation
where would you start from;
what steps will you follow;
what will you do at each step.
4
Ontology Development Methodologies
Historically,there have been several different methodologies and methods
Depend on the scale and stakeholders
Analogy with software engineering
methodologies
General agreement on:
Main steps needed
Need for iterations
5
Methodology vs Method
Methodology:
“comprehensive, integrated series of techniques creating a general systems theory of how class of thought- intensive work ought be performed”
Method:
“orderly process or procedure used in engineering
of a product or performing a service” definitions from [IEEE,1990; IEEE 1995]
6
Terminology
Competency Questions
Questions the Ontology must be able to answer
Inter-Lingua
An artificial language, devised for machine translation, that makes explicit the distinctions necessary for successful translation into a target language, even where they are not present in the source language. [en.oxforddictionaries.com] (other meanings)
7
Additional Reading
Six papers on minerva
1. Uschold and King Method
2. Gru ̈ninger and Fox Methodology
3. Methontology – Building Chemical Ontology
4. Kanga: Ordnance Survey ontology development
5. NEON Project
6. Niemann, Mochol and Tolksdorf: Hotel Ontology
Read and identify the main ideas and issues in these. You are not expected to study every small detail.
8
Uschold & King Method
http://www.aiai.ed.ac.uk/project/enterprise/enterprise/ontology.html
Why built Ont. Intended Uses Potential Users Scope
Key concepts and relationships Textual definitions
Other terms that refer to them Bottom-up/top-down/middle-out
Representation of key concepts Coding of the knowledge
Integrating existing ontologies
Specifications Competency Q Real world uses
Conventions Modelling decisions
9
Grüninger & Fox Methodology
http://www.eil.utoronto.ca/enterprise-modelling/tove/
Illustrates the use of the ontology
Source for requirements formulation
Questions to be answered by the ontology once it is ready
Define main objects
Define main predicates
Formulate the competency questions with the formal aparatus
Define facts that hold in the domain in a logical way
Formulate and prove claims regarding ontology completeness
10
METHONTOLOGY
http://oa.upm.es/5466/1/Building_a_Chemical_Ontology.pdf
11
METHONTOLOGY
http://oa.upm.es/5466/1/Building_a_Chemical_Ontology.pdf
-Purpose -Scope -Granularity -Sources used (interviews with experts; documentation)
-Identify main concepts and relationships
– Knowledge glossary
-Define class hierarchy
– Describe axioms & rules
-Coding -Validation
using an
ontology -Reuse building
tool
12
KANGA
https://link.springer.com/content/pdf/10.1007%2F978-3-540-88564-1_1.pdf Identify scope and purpose
Verify the ontology Use different techniques
Define concepts, relationships, axioms in a controlled natural language then do the logical coding
Identify knowledge sources Identify core concepts/relationships
Glossary of key concepts Glossary of key relationships
13
Putting it ALL Together: NEON
http://www.neon-project.org/web-content/images/Publications/neon_2008_d5.4.1.pdf
-Purpose -Scope -Granularity -Knowledge sources
– Other ont. -Intended uses -Intended users -Scenarios -Competency questions
Ontology requirements
-Main concepts & relationships -Top-down / bottom-up / middle-out -Seed terms -Core/secondary concepts -Glossaries
Logical conceptualisation Ontology
Ontology
-Coding in the knowledge representation framework chosen
-Documentation -Validation -Application -Reuse
Reports and Descriptions
14
NEON Ontology Lyifecycle
http://www.neon-project.org/web-content/images/Publications/neon_2008_d5.4.1.pdf
15
Summary
Ontology Development is an iterative process involving several main steps, producing:
Requirements specifications
Ontology conceptualisation
Logical ontology
Documentation and reports
Ontology Development involves various stake holders:
Ontology engineers
Domain experts
Application developers
Application users
There are several Ontology Development Methodologies
Historically developed at different times and contexts
Usually combined and adapted for the specific case
16